Grafana OnCall Quiz
A quiz to test your understanding of Grafana OnCall.
- Which is NOT a key feature of Grafana OnCall?
- A) On-call schedule management
- B) Escalation chain configuration
- C) Metric collection and storage
- D) ChatOps integration (Slack, Teams)
Show Answer
Answer: C) Metric collection and storage
Explanation: Grafana OnCall is an on-call management and incident response tool that provides the following features:
- On-call schedule management (rotations, overrides)
- Escalation chain configuration
- Alert grouping and routing
- ChatOps integration (Slack, MS Teams, Telegram)
- Mobile app notifications
Metric collection and storage is the role of other tools like Prometheus or Grafana Mimir. OnCall receives and processes alerts generated from these tools.
- What is the role of the
waittype in Grafana OnCall's escalation policy?- A) Wait for data collection before sending alerts
- B) Wait before proceeding to the next escalation step
- C) Wait for user response then auto-resolve
- D) Wait for alert grouping
Show Answer
Answer: B) Wait before proceeding to the next escalation step
Explanation: In escalation chains, the wait type sets a wait time between the current step and the next step. For example:
- Step 1: Notify current on-call responder
- Step 2: wait 900 seconds (15 minutes)
- Step 3: If no response, notify secondary responder
This gives the primary responder time to respond, and escalation only proceeds if there is no response.
- What is an "Override" in Grafana OnCall's on-call schedule?
- A) Completely deleting and recreating the schedule
- B) Temporarily changing the responder for a specific period in the existing schedule
- C) Changing the schedule's time zone
- D) Modifying the rotation cycle
Show Answer
Answer: B) Temporarily changing the responder for a specific period in the existing schedule
Explanation: Override is a feature that temporarily changes the responder for a specific period in a regular on-call schedule. Main use cases:
- Responder replacement due to vacation
- Temporary change due to emergency situations
- Temporary swap due to training/meetings
Override designates a different responder for a specific period while maintaining the existing schedule.
- What method is used when integrating Grafana OnCall with Alertmanager?
- A) Alertmanager directly collects OnCall's metrics
- B) Send alerts to OnCall via Alertmanager's webhook_configs
- C) OnCall periodically polls Alertmanager's API
- D) Both systems share a database
Show Answer
Answer: B) Send alerts to OnCall via Alertmanager's webhook_configs
Explanation: Alertmanager and Grafana OnCall integration is done through webhooks:
receivers:
- name: 'grafana-oncall'
webhook_configs:
- url: 'https://oncall.example.com/api/v1/webhook/<integration-id>/'
send_resolved: trueWhen Alertmanager fires an alert, it sends an HTTP POST request to the configured webhook URL, and OnCall receives and processes it.
- What is the main purpose of alert grouping in Grafana OnCall?
- A) Sort alerts by time
- B) Group related alerts into one to reduce alert fatigue
- C) Classify alerts by severity
- D) Automatically delete duplicate alerts
Show Answer
Answer: B) Group related alerts into one to reduce alert fatigue
Explanation: Alert grouping manages multiple alerts caused by the same problem as a single group. For example, if multiple pod alerts occur due to a node failure, they are grouped together so the responder receives one grouped alert instead of dozens of individual alerts. A grouping key (e.g., alertname + namespace) is defined to determine which alerts are grouped together.
- What happens when the
importantflag is true fornotify_on_call_from_schedulein Grafana OnCall's escalation policy?- A) Alert is marked as highest priority
- B) Alert is sent via all configured channels (phone, SMS, push, etc.)
- C) Escalation chain is skipped and alert is sent immediately to supervisor
- D) Alert is permanently stored
Show Answer
Answer: B) Alert is sent via all configured channels (phone, SMS, push, etc.)
Explanation: The meaning of the important flag:
important: true: Send alert via all notification channels configured by the user (phone, SMS, mobile push, Slack, etc.)important: false: Send alert via default channels only (e.g., Slack)
This allows adjusting alert intensity based on severity. Critical alerts can be set to important=true to include phone/SMS, while Warning can be set to important=false to use only Slack.
- Which is NOT a command available when using Slack integration with Grafana OnCall?
- A) /oncall ack (acknowledge alert)
- B) /oncall resolve (resolve alert)
- C) /oncall deploy (execute deployment)
- D) /oncall silence 2h (silence for 2 hours)
Show Answer
Answer: C) /oncall deploy (execute deployment)
Explanation: Grafana OnCall Slack commands:
/oncall- Check current on-call responder/oncall schedule- View schedule/oncall ack- Acknowledge alert/oncall resolve- Resolve alert/oncall silence 2h- Silence for 2 hours/oncall unsilence- Remove silence/oncall escalate- Escalate alert
Deployment execution is not an OnCall feature. OnCall focuses on alert management and on-call management.
- What is NOT an advantage of Grafana OnCall compared to PagerDuty/OpsGenie?
- A) Open source and can be self-hosted
- B) Native integration with Grafana stack
- C) Support for 700+ integrations
- D) Free to use (OSS version)
Show Answer
Answer: C) Support for 700+ integrations
Explanation: 700+ integrations is an advantage of PagerDuty. Comparison:
- Grafana OnCall: 30+ integrations, open source, self-hosting possible, Grafana native integration, free (OSS)
- PagerDuty: 700+ integrations, SaaS only, advanced analytics/reporting, AIOps features
- OpsGenie: 200+ integrations, SaaS only, Atlassian ecosystem integration
OnCall is a cost-effective choice for environments using the Grafana stack, but PagerDuty may be more suitable when diverse external system integrations are needed.
- What is the recommended configuration for high availability in Grafana OnCall production deployment?
- A) A single instance is sufficient
- B) Increase API server and Celery worker replica counts and use external PostgreSQL/Redis
- C) Distributed deployment across multiple clusters
- D) Only add read-only replicas
Show Answer
Answer: B) Increase API server and Celery worker replica counts and use external PostgreSQL/Redis
Explanation: Grafana OnCall Production HA configuration:
- API server: 3+ replicas, Pod Anti-Affinity settings
- Celery workers: 3+ replicas, Pod Anti-Affinity settings
- PostgreSQL: Use external managed DB (AWS RDS, etc.)
- Redis: Use external managed Redis (AWS ElastiCache, etc.)
This configuration eliminates single points of failure, allowing the service to continue operating even when individual components fail.
- What is the main purpose of setting up routes in Grafana OnCall?
- A) Network traffic distribution
- B) Apply different escalation chains based on alert conditions
- C) Database query optimization
- D) User authentication path configuration
Show Answer
Answer: B) Apply different escalation chains based on alert conditions
Explanation: Routes connect incoming alerts to appropriate escalation chains based on their attributes (labels, severity, team, etc.):
severity=critical-> Critical escalation chain (includes phone/SMS)team=infra-> Infrastructure team escalation chainnamespace=production-> Production on-call schedule
Routing rules are defined using regular expressions and match based on the alert payload content. This allows applying appropriate responders and escalation policies for different alert types.